Due to the ever more progressive miniaturization of portable electronic equipment, the demand for ever smaller and lighter secondary batteries which serve as an energy source for such equipment has risen at tremendous speed in recent years.
It has been found that high energy densities can be achieved in particular with lithium secondary batteries. This type of secondary battery is distinguished by a positive electrode, the active material of which can reversibly embed and release lithium ions. The embedding of lithium ions takes place in certain layers of the crystal lattice of the active material and proceeds all the more efficiently the fewer defects there are in the layer structure of the crystal lattice. LiCoO2 in particular is employed successfully as the active material. LiCoO2 is distinguished by a very stable crystal structure, and secondary batteries in which the LiCoO2 achieves a discharge capacity of about 150 mAh/g at an average potential of 3.7 V can be produced therefrom. On the basis of the high costs for cobalt, alternative active materials which, where possible, also render even better discharge capacities are being intensively sought. LiNiO2 has acquired great interest in this respect, since it has been found that when LiNiO2 is employed, discharge capacities of more than 200 mAh/g can be achieved. However, LiNiO2 is significantly more susceptible to the development of defects in the crystal structure than LiCoO2, so that a sufficiently high cycle stability does not result when LiNiO2 is employed.
It has therefore been proposed to employ as the active material LiNiO2 which contains further metals in addition to nickel. In particular, doping with cobalt and aluminium has proved advantageous. Thus, doping with cobalt and aluminium increases the heat stability. However, aluminium does not contribute towards the discharge capacity and should therefore be added in such a small amount that although the desired increase in stability is achieved, an unnecessary increase in weight is avoided. In this context, it is decisive that the aluminium is homogeneously distributed in the active material.
In the preparation of the active materials, typically a lithium compound is mixed with hydroxides of the desired metal components and the mixture is calcined. The homogeneity of the distribution of the metal components in the crystal lattice of the active material depends considerably on how successfully a homogeneous distribution of the constituents is to be already achieved in the mixture to be calcined. It has been found that conventional mixing and grinding of the lithium compound, nickel component, cobalt component and aluminium component is not suitable for reliably mixing large amounts homogeneously.
Attempts have therefore already been made to employ mixed metal hydroxides in which the metals are already homogeneously distributed, instead of the simple hydroxides of the desired metal components. Thus, mixed nickel/cobalt hydroxides in which the metals are homogeneously distributed in a solid solution are obtained by co-precipitation. On the other hand, the co-precipitation of mixed metal hydroxides which comprise nickel and aluminium presents difficulties, since as the amount of aluminium added increases, the filterability of the co-precipitate formed decreases, and the removal of the anions of the metal salts employed during the co-precipitation also becomes problematic.
JP 11-016752 therefore proposes starting from nickel hydroxide or a nickel/cobalt hydroxide and depositing aluminium hydroxide thereon. For this, an alkali metal aluminate is first dissolved in a suspension containing the optionally cobalt-containing nickel hydroxide. The then strongly alkaline suspension is neutralized by dropwise addition of an acid, as a result of which aluminium hydroxide is formed, which precipitates out, with mixing and adsorption on the surface of the nickel hydroxide. The suspension is stirred intensively during this procedure. After conclusion of the addition of an acid, the mixture is stirred for about a further 30 minutes and the precipitate is then filtered off and dried. The precipitate obtained in this way is redispersed in water, washed and finally dried again. Due to the long dwell time in the precipitation reactor, the high concentration of solid in the suspension and the intensive stirring required, the particles are exposed to severe friction, so that there is the danger that aluminium hydroxide particles which have already been adsorbed are partly abraded away from the surface of the nickel hydroxide. The long dwell time in the precipitation reactor and the very slow change in pH by dropwise addition of an acid furthermore cause aluminium hydroxide particles of different crystal structure or morphology to form. The saturation concentration for the aluminium compound is reached and exceeded very slowly, so that comparatively large aluminium hydroxide particles of high crystallinity form. This has the consequence that the aluminium can diffuse poorly into the core of the mixed metal hydroxide particles during the subsequent thermal reaction of mixed metal hydroxide and lithium compound. There is the danger of the formation of undesirable lithium aluminate phases, such as Li5AlO4 and LiAlO2, and a uniformly homogeneous distribution of aluminium in the material formed is not ensured.
JP 2001-106534 A1 also discloses mixed metal hydroxides which are employed as a starting material for the preparation of active material for the positive electrode of a secondary battery. Co-precipitated nickel/cobalt hydroxide is again used as the starting material, and is coated with aluminium hydroxide. Coating is carried out in a reaction tower by addition of an aluminium nitrate solution. The pH is adjusted to weakly basic, so that the aluminium nitrate is converted into aluminium hydroxide, which is deposited on the surface of the nickel/cobalt hydroxide. Coating is again carried out with stirring, so that in this procedure also the coated particles are exposed to severe mechanical stress and there is the danger of detachment of the aluminium hydroxide from the surface of the nickel/cobalt hydroxide. The reaction conditions in a stirred reaction tower in turn cause a long dwell time and the formation of aluminium hydroxide particles of high crystallinity. In the material obtained in this way, the aluminium thus also can diffuse poorly into the core of the mixed metal hydroxide particles during the subsequent thermal reaction of mixed metal hydroxide and lithium compound, so that a homogeneous distribution of the aluminium in the material formed is not ensured.